This is one of the most exciting seasons in the history of the English Premier League as five teams have a chance of winning the title. Arsenal, Chelsea, Manchester City, Manchester United and Tottenham Hotspur all have a shot at being champions. So who are the likely winners?
Premier League Trophy (Source: Sky Sports)
It has been argued by the Manchester United manager Sir Alex Ferguson that his team are historically stronger in the second half of the season. That surely implies that the current league leaders are well on the way to a record 19th league title. The first step of this analysis is to test if this hypothesis is in fact true. Using data from the last five complete seasons, we can compare the share of points gained in the first half of the season with the second half. Extrapolation of data to gain valuable insight is an important element of any Operational Research project. The data used in this article is from worldfootball.net.
Table 1: Manchester United – Points earned during the season as a percentage of the total points gained
Table 1 illustrates how the total points tally gained by Manchester United for the last five seasons is distributed between the first 19 games and the final 19 games (there are 38 games in one season). It can be concluded that the points gained in both halves of the season are more or less equal. So while Manchester United might be appear to be stronger in the second half of the season, they actually gain there or thereabouts the same number of points as they did in the first half of the season. Forecasting is an important and very useful technique that is used by Operational Research consultants. The points distribution between the two halves of the season can be used as a means of forecasting the final points tally for the five teams in questions.
Table 2: Points gained during the first half of the season as a percentage of the total points gained
Table 2 shows the average points gained in the first half of the season for the five teams in question using the data for the last five seasons. Like Manchester United, the points spread between the two halves of the season are relatively equal. Tottenham Hotspur is slightly different in that historically they gain a lot more of their points in the second half of the season. To forecast the final points total for each of the five teams for the 2010/2011 season, the methodology is simple. All we have to do is to take the points gained after 19 games and divide it by the average percentage of points gained in the first 19 games of the season for the last five years. The results are shown in Table 3.
Table 3: Projected Final Points Tally
And the champions are….Manchester United. What is quite evident based on projections is that the season might not be as close as many football fans believe it will be. The range of points for the five teams after 19 games was 8 points and this is projected to increase to 18 points with Manchester United ending up being the runaway champions. Whilst simple methods can be used to gain initial insight into many business problems, there are many more sophisticated methods for forecasting. For example, we can give more weighting to the recent seasons; capture aspects such as recent form and factor in future known events. In this case it could be that form could drop for certain teams as a result of involvement in other competitions. Forecasting techniques are chosen depending on the business issue at hand. Based on the last five seasons worth of data, a team would have needed on average 85 points to win the title.
Table 4: Second half of season average points per game
Table 4 shows for each of the teams the projected average points per game for the second half of the season based on each team’s projected final points tally. The average points per game required to reach the magic number of 85 is also reported. As expected, Manchester United are the only team that are on track to achieve the title. So whilst the media would like us all to believe that we set for an exciting second half to the season, this might in fact not be the case.